Few instruments exist for assessing writing and those that do focus on the product and not the process of writing. Since many school- age children are referred for assessment of writing disabilities, this research has been designed to develop a theory-based instrument for assessing components of the writing process (specific aim 1) and to establish criteria for writing disabilities that can be used for subject selection in future research on writing disabilities (specific aim 2). Study 1 will investigate the neuropsychological components that may constrain writing in grades 1-3. Study 2 will investigate linguistic components that may constrain writing in grades 4-6. Study 3 will investigate cognitive components that may constrain writing in grades 7-9. One hundred children (50 girls, 50 boys) at each of three grade levels (in an unreferred sample representative of ethnic and socioeconomic groups in the U.S. population) will participate in each study. In Study 1 concurrent measures of 3 predictor variables (letter retrieval, finger tasks used to elicit neurological soft signs, and visual-motor integration) and 3 criterion variables (copying words, spelling words, and writing a paragraph) will be administered; multiple regression will identify the set of ordered variables that accounts for the most variance in each criterion variable. In Study 2, which is an extension of pilot work, 4 measures of verbal intelligence, 4 measures of oral language, and 4 measures of reading (predictor variables) and 4 measures of writing (criterion variable) will be administered concurrently; multiple regression will identify the set of ordered predictor variables that accounts for the most variance in the criterion variable and structural equation modeling (EQS, Bentler, 1985) will be used to test a structural model linking the 3 predictor factors (underlying measures of verbal intelligence, oral language, and reading) and the 1 criterion factor (underlying writing measures). In Study 3 a planning task, translating task, and reviewing/revising task will be administered to measure each component process in the model of writing proposed by Hayes and Flower (1980); the relationship between a variety of dependent measures on these tasks and the judged quality of essays produced during translating and revising will be evaluated. This research may contribute to improved clinical assessment of writing disabilities and prevention of mental health problems related to chronic writing disabilities in school-aged children.